Biosurveillance: Capacity and Utility
Syndromic surveillance (SyS) is a biosurveillance methodology that integrates new information technologies, electronic health record data, and principles of epidemiology and biostatistics. The data are collected from hospitals and public health departments and transmitted to public health agencies in near-real time (at least every 24 hours). The data reported are categorized as a “syndrome” on the basis of a collection of symptoms or other variables. The aggregate number of cases reported within syndromes reported is analyzed by trained practitioners (ie, epidemiologists) to prevent or mitigate potential outbreaks and assess the overall health of a community. SyS is an important tool that transcends the traditional use of detecting and monitoring bioterrorism events. The Centers for Medicare & Medicaid Services has included SyS reporting as a requirement for eligible hospitals for stage 2 Meaningful Use, which emphasizes the importance of the technology on a national level.1 This technology can be a useful tool when monitoring population health issues such as drug-related hospitalizations, natural disaster–related illness/injury, and noncommunicable diseases/chronic disease. Public health activities and response can be further improved by the utility of SyS technology.
Purpose of the 2015 Biosurveillance Needs Assessment
In 2014, the National Association of County & City Health Officials (NACCHO) conducted a survey of 500 local health departments (LHDs) to further analyze the results of the NACCHO 2013 National Profile of Local Health Departments, which found that 62% of all LHDs surveyed had used some form of an electronic SyS system. This overwhelming response prompted staff at the Centers for Disease Control and Prevention (CDC), NACCHO, and the International Society for Disease Surveillance (ISDS) to collaborate on the execution and assessment of the SyS capabilities of LHDs across the United States—Syndromic Surveillance Practice at Local Health Departments in the United States: Results from the 2015 NACCHO Biosurveillance Needs Assessment (2015 BNAS). The purpose of the study was to understand the nationwide management and use of electronic SyS practice among a sample of LHDs and to identify gaps and barriers to improve both individual and organizational capabilities in biosurveillance, as measured by SyS practice. The information from the survey is helpful in guiding the efforts of CDC, NACCHO, and ISDS, as well as other relevant partner organizations, in determining the public health surveillance needs of LHDs across the country. Potential activities include addressing specific technical assistance needs of novice, intermediate, and expert users of SyS, as well as providing resources in the form of webinars, online distance learning opportunities, and written resources.
Study population and sampling
The survey was distributed to a random sample of 500 LHDs from the NACCHO database. The sample criteria used to categorize the LHDs were SyS status and size of population served. The SyS status was determined by the LHD response to the 2013 National Profile of Local Health Departments. The criteria involved a random selection of LHDs—60% of individuals in the sample were from those who responded “yes,” 20% were from those who responded “no,” and another 20% were from nonrespondents to the previous NACCHO survey. The sample was stratified by the size of population within each group (<50 000, 50 000-499 999, and ≥500 000). LHDs from the “large” population size category were oversampled because they represented a smaller proportion of all LHDs to ensure a sufficient number of responses.
The 2015 report involved a mixed-methods approach using both quantitative and qualitative analyses. NACCHO formed the ad hoc Biosurveillance Needs Assessment Survey (BNAS) workgroup, composed of SyS experts from local, state, and federal SyS practice to provide input and feedback into the development of the survey instrument and key informant interview guide. During monthly calls, the NACCHO/ISDS project team and workgroup developed a framework that established the assumptions and objectives for the survey. A pilot of the survey was distributed in February 2015 to 5 LHDs. On the basis of the pilot and the feedback from the workgroup, the survey was revised and distributed in March 2015.
Qualitative data from key informant interviews provided a deeper understanding of the use of SyS and LHD biosurveillance needs. The BNAS workgroup provided input and feedback on the key informant interview guide developed by project staff. The participants of the key informant interviews represented jurisdictions of different sizes, varying use of SyS, and whether their system was managed directly by their LHD or by another entity (ie, a state health department). The interview guide included questions regarding the surveillance practice of the informants. Questions were tailored on the basis of the informant's survey responses. ISDS conducted 5 interviews by phone. Notes were taken and reviewed by ISDS staff who came to a consensus on salient points.
An electronic link was sent by ISDS along with an introduction letter from NACCHO to the sample list, indicating the target responder. The target responder in a jurisdiction that uses SyS was the individual most familiar with the use of the technology and/or with an understanding of the concept of electronic emergency department SyS. The target responder in a jurisdiction that does not use SyS was an individual whose responsibilities included epidemiology and disease surveillance. The survey respondent was welcome to consult with other colleagues. The instructions explained that the answers should represent the responder's personal experiences and perceptions and are not necessarily indicative of the entire health department. The survey was administered between March 2, 2015, and April 27, 2015, with 3 automatic reminders issued.
If respondents did not answer whether they had access to SyS or whether their LHD managed their own system, the response was removed from the analysis. For any duplicated responses submitted by an LHD, the later response (identified by the date stamp) was only included. Incomplete surveys were included, and the questions were analyzed individually with values for the number of response indicated. The results that are presented in the 2015 BNAS report are by the number of respondents and are not weighted averages to allow for the best level of input from SyS practitioners. The responses were not weighted by LHDs on the basis of their population served but were analyzed by the percentage of respondents. While the results of the survey provide a snapshot for the technical needs of LHDs engaged in SyS across the United States, the responses are not meant to be representative of national practice. However, for the purposes of this article, all figures have been weighted to account for the sampling plan that oversampled responses from large health departments. This is to reduce the occurrence of any sampling bias.
LHD use of SyS
Figure 1 represents the percentage of respondents who have access to SyS (48%) and do not have access to SyS (52%). This shows that a little less than half of the LHDs that completed this survey have access to this technology. Figure 2 represents LHDs that have access or no access to SyS by population size (small, <50 000; medium, 50 000-499 999; large, >500 000). These results indicate that generally smaller LHDs have more instances of not having access to SyS (60%) compared with medium-sized LHDs (44%) and large LHDs (19%). However, access to SyS is more prevalent among the large LHDs sampled (81%) compared with medium-sized LHDs (56%) and small LHDs (40%).
Figure 3 shows the type of SyS applications used by LHDs. The ESSENCE application, developed and licensed by the Johns Hopkins University Applied Physics Laboratory (JHU-APL), is used by 27% of LHD respondents and represented the highest usage reported. The second highest option selected is “other” denoted by about 25% of respondents. Most of the “other” responses indicated that they use a system developed by the state health department. The third highest option selected is locally developed system at 19%.
Utility of SyS for public health surveillance
Respondents were asked several questions to understand how LHDs use SyS. Respondents answered these questions only if they had indicated their LHD has access to SyS data or information. In response to the question “Is SyS incorporated into LHD response protocols?” a majority of respondents (64% of LHDs with SyS) reported that their LHD incorporates SyS into their response protocols; 22% said that SyS is not part of their response protocols, and 13% were not sure. During the key informant interview portion of the survey, a respondent indicated that response protocols are handled at the state level. This result exemplifies the need for greater collaboration and information sharing between state and LHDs with regard to SyS.
Survey respondents were presented with 16 public health areas where using SyS was reported on the basis of the literature from ISDS conference abstracts and input provided by the BNAS workgroup. They were asked to identify which areas were monitored by their LHD using any surveillance method, including SyS. In a second question, they were asked to identify which of the 16 they monitor using only SyS. The responses shown in Figure 4 display the comparison between “any surveillance” method versus “SyS.” Respondents had the option to write-in a response; examples include weather-related injury/illness and carbon monoxide poisoning. A key informant noted that once users are comfortable with using SyS, they could use the system to monitor new areas. The top categories monitored (>50% of respondents) using any surveillance method include foodborne illness (95%), influenza-like illness (ILI) (91%), notifiable diseases (88%), gastrointestinal illness (80%), vector-borne disease (79%), animal bites (78%), and bioterrorism agents (68%). The least monitored (<20%) using any surveillance method include suicide and self-inflicted injury (19%), alcohol-related injuries (14%), mental health (8%), and wildfire effects (6%). The top (>50% of respondents) categories monitored using SyS include only ILI (74%), gastrointestinal illness (67%), foodborne illness (57%), and notifiable diseases (52%). The least monitored (<20% of respondents) grows to a much wider pool—poisoning (14%), noncommunicable diseases/chronic disease (11%), drug-related hospitalizations (10%), injuries (9%), suicide and self-inflicted injury (8%), alcohol-related injuries (7%), mental health (5%), and wildfire effects (6%).
Figure 5 shows the percentage of LHD respondents who use SyS to monitor health issues by level of SyS system management. SyS system management is divided by whether the LHD manages its own system or does not manage its own system. The results show that LHDs that manage their own system use it more frequently than LHDs that have to access data from an external system. LHDs that manage their own system could have more control over what they use SyS to monitor and potentially use it for a variety of syndromes. Generally, LHDs that can house their own system tend to represent those from the large population category (500 000+).
LHD use of SyS and workforce capacity
In assessing LHD use of SyS, it is important to assess the workforce capacity of LHDs. Figure 6 is in response to the question, “Why do LHDs not use SyS data and information?” The top 3 reasons indicated are monetary resources (60%), staff resources (58%), and technical capabilities (41%). A key informant interviewer included the following response that exemplifies how important monetary resources are to LHDs.
Lack of funding, which in turn means staff time, is by far the biggest barrier to our department moving in this direction. We will not be able to use much or any of the above tools without funding or dropping other priority work.
Funding for LHDs is critical not only for managing and using SyS systems but also for training and retaining staff in SyS. All of these aspects are key to furthering the SyS capabilities of LHDs. Generally, smaller LHDs (population size <50 000) are in greater need of funding to use SyS and retain staff who have had some background related to SyS, such as epidemiology. A key informant interview alluded to that as well: the individual stated that a lack of public health funding overall and specifically low wages for epidemiologists creates “a lot of turnover.” The key informant also noted that training of staff is very critical to the overall workforce capacity of the LHD; the informant stated that “even if the state did share data, most of the jurisdictions don't have qualified epidemiologists to use the data.”
The statement by the key informant is extremely relevant to the overall technical assistance needs of LHDs. SyS practitioners come to their health department with a diverse range of experience. Some of these practitioners are more advanced than others. The responses from practitioners to this survey indicate that “one size will not fit all for both the content and the mechanisms for technical assistance in biosurveillance.” Survey respondents were asked, “What type of activities would help LHDs that do not currently use SyS to consider using it?” The responses to the question are indicated in Figure 7 and are categorized by those who use SyS and those who do not use SyS. Out of the respondents who use SyS, webinars (78%), written resources (47%), and online distance learning opportunities (42%) are the top 3 choices. Of the respondents who do not use SyS, information about funding opportunities (54%), webinars (48%), and technical assistance (41%) are the top 3 choices. Webinars is the common choice between the 2 respondent groups. Figure 8 examines how survey respondents prefer to receive information about events and new practices in SyS. The results indicate that e-mail and webinars (80% and 49%, respectively) are the top choices. This further validates the results from Figure 7 regarding activities to boost SyS usage at the respondent's health department.
Discussion and Conclusion
The results from the BNAS report provide a picture of the utility of SyS practice at LHDs across the country. The results of the survey can further guide technical assistance activities to improve the overall capacity of LHDs to engage in SyS. The results indicated that 64% of LHDs that use SyS incorporated it into their response protocol. This result shows the value of SyS among LHDs that use the technology especially to respond to a potential public health crisis. Because SyS collects data in near-real time, a potential outbreak impacting a local community could be better mitigated if LHDs have the tools and resources available to respond with general oversight from the state health department. It is important to provide LHDs that have either no SyS system in place or a novice system and/or user of the technology with tools or resources that can build the capacity within the health department. The BNAS report categorized LHDs into 3 areas related to their SyS capabilities: (1) LHDs that do not currently have access to SyS, (2) LHDs with a basic level of SyS practice (which generally includes LHDs that do not manage their own system, and (3) LHDs with more advanced levels of SyS practice (including LHDs that do not manage their own system, represent populations over 50 000, and have SyS professionals with greater than 3 years of experience).
For all 3 categories, it is important to first build the utility and value of SyS among LHDs. This can be done through informational webinars that give an overview of the technology and are focused on a particular category (ie, novice or intermediate user). Also, it is important to explain the value of technology. Figures 4 and 5 delve into what health issues (ie, ILI) respondents monitor using SyS or any surveillance method as well as what health issues they monitor using their own system or another system (ie, state-run system). The distinction of whether an LHD monitors a health issue with a system they manage or do not manage is important because the system may not be as user friendly if it is not made for the specific community. The information in these figures is important to address because it showcases the wide range of syndromes that can be monitored using SyS. It also shows the effect of an LHD managing its own system versus using a state-run system. Issues regarding whether an LHD is able to see data from surrounding communities can affect how to respond to potential public health threats. Sometimes LHDs do not have access to state-wide data because of data-sharing laws. One way to combat this is to provide a data-sharing agreement template for LHDs to gain access to data from hospitals, health departments, etc., within and outside their jurisdictions.
Figures 6 and 7 focus on barriers to SyS practice and potential activities that would help improve or implement practice respectively. The top 3 reasons indicated as barriers are monetary resources, staff resources, and technical capabilities. These barriers affect LHDs at varying degrees. Smaller LHDs generally do not have the technology as well as the staff capacity and/or expertise to implement SyS. Larger health departments struggle with the same issues but in a different way. Staff that come into health departments have varying expertise and background in SyS, and some may be more experienced than others. Funding to improve the current state of SyS at medium to large LHDs is generally contingent on funding from state and federal governments. In these cases it is often best to develop a community of practice and also engage LHDs to reach out to those in their surrounding communities to share best practices and stories from the field in addition to applying for more funding opportunities. Online distance learning opportunities through webinars are necessary? important to further developing practitioners in the field and introducing novice users. As indicated by the results, webinars, written resources, and online training opportunities are very important to those who use SyS and those activities should be provided regularly by associations that conduct capacity-building activities for public health agencies. Respondents who do not use SyS are very interested in information about funding opportunities. Funding opportunities can be especially useful to LHDs with little to no experience with SyS because it is a large cost for health departments to invest in the needed infrastructure (ie, personnel and training) and technology. This information is critical to associations that represent health departments that need to focus their activities to better disseminate trainings, written resources, and funding opportunities. SyS is a technology that can be highly useful to health departments to boost the overall health of their communities if there is the right combination of accessibility, use, and training of the technology by the end user.Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.